499
British Journal of Health Psychology (2016), 21, 499–514
© 2015 The Authors. British Journal of Health Psychology published by
John Wiley & Sons Ltd on behalf of British Psychological Society
www.wileyonlinelibrary.com
Significant other behavioural responses and
patient chronic fatigue syndrome symptom
fluctuations in the context of daily life: An
experience sampling study
Rebecca Band1,2*, Christine Barrowclough1, Richard Emsley3,
Matthew Machin4 and Alison J. Wearden1
1
School of Psychological Sciences & Manchester Centre for Health Psychology,
University of Manchester, UK
2
Centre for Applications of Health Psychology, University of Southampton, UK
3
Centre for Biostatistics, Institute of Population Health, University of Manchester, UK
4
Centre for Health Informatics, Institute of Population Health, University of
Manchester, UK
Objective. Significant other responses to patients’ symptoms are important for patient
illness outcomes in chronic fatigue syndrome (CFS/ME); negative responses have been
associated with increased patient depression, whilst increased disability and fatigue have
been associated with solicitous significant other responses. The current study aimed to
examine the relationship between significant other responses and patient outcomes
within the context of daily life.
Design. Experience Sampling Methodology (ESM).
Method. Twenty-three patients with CFS/ME and their significant others were
recruited from specialist CFS/ME services. Sixty momentary assessments, delivered
using individual San Francisco Android Smartphones, were conducted over a period of
6 days. All participants reported on affect, dyadic contact, and significant other responses
to the patient. Patients reported on symptom severity, disability, and activity management
strategies.
Results. Negative significant other responses were associated with increased patient
symptom severity and distress reported at the same momentary assessment; there was
evidence of a potentially mediating role of concurrent distress on symptom severity.
Patient-perceived solicitous responses were associated with reduced patient activity and
disability reported at the same momentary assessment. Lagged analyses indicate that
momentary associations between significant other responses and patient outcomes are
largely transitory; significant other responses were not associated with any of the patient
outcomes at the subsequent assessment.
Conclusion. The results indicate that significant other responses are important
influences on the day-to-day experience of CFS/ME. Further research examining patient
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly cited.
*Correspondence should be addressed to Rebecca Band, Academic Unit of Psychology, University of Southampton, Shackleton
Building, Highfield Campus, Southampton SO17 1BJ, UK (email: [email protected]).
DOI:10.1111/bjhp.12179
500
Rebecca Band et al.
outcomes in association with specific significant other behavioural responses is
warranted and future interventions that target such significant other behaviours may
be beneficial.
Statement of contribution
What is already known on this subject?
The existing literature has identified that significant other responses are important with respect
to patient outcomes in CFS/ME. In particular, when examined cross-sectionally and longitudinally, negative and solicitous significant other responses are associated with poorer illness
outcomes. This study is the first to examine the momentary associations between negative and
solicitous responses, as reported by the patient and significant other, and patient-reported
outcomes. An ESM paradigm was used to assess these temporal relationships within the context
of participants’ daily life.
What does this study add?
Negative responses were associated with increased momentary patient distress and symptoms.
Perceived solicitousness was associated with activity limitation but less perceived disability.
The impact of significant other responses on patient outcomes was found to be transitory.
Chronic fatigue syndrome or myalgic encephalomyelitis (CFS/ME) is a symptomatically
defined condition, characterized by severe fatigue and pain (Fukuda et al., 1994). Current
explanatory models suggest that patients’ cognitive, behavioural, and affective responses to
symptoms are important for symptom perpetuation (Deary, Chalder, & Sharpe, 2007;
Surawy, Hackmann, Hawton, & Sharpe, 1995). Interpersonal relationships may influence
these maintaining factors (e.g., by reinforcing thinking patterns or illness management
behaviour, or by providing a source of support or stress for the patient; Band, Barrowclough,
& Wearden, 2014; Deary et al., 2007), and interactions with significant others have been
highlighted as important in the patient illness experience (Dickson, Knussen, & Flowers,
2007). Cognitive behaviour therapy and graded exercise therapy, both of which encourage
gradual increases in daily activity levels, are effective treatments for CFS/ME (Castell,
Kazantzis, & Moss-Morris, 2011; White et al., 2011). To date, there is little research on the
impact that significant others may have on patients’ responses to these treatments.
A recent review of the literature examining significant other responses to CFS/ME
identified two types of responses associated with patient-reported CFS/ME outcomes
(Band, Wearden, & Barrowclough, 2015). Significant other negative responses, such
expressing frustration at the patient (Kerns & Rosenberg, 1995), have been found to be
associated with increased patient depression (Romano, Jensen, Schmaling, Hops, &
Buchwald, 2009; White, Lehman, Hemphill, Mandel, & Lehman, 2006). In turn, patient
depression has been associated with poorer long-term patient illness outcomes and
responses to treatment (Bentall, Powell, Nye, & Edwards, 2002; Wearden, Dunn,
Dowrick, & Morriss, 2012). The review proposed that elevated levels of patient
depression may mediate an association between negative significant other responses and
increased fatigue (Band et al., 2015), and this association has indeed been observed
longitudinally (Band et al., 2014). The second type of response proposed in the review
was solicitous significant other responses, such as encouraging patients to rest or doing
tasks on their behalf (Cordingley, Wearden, Appleby, & Fisher, 2001; Kerns & Rosenberg,
1995). In several studies, solicitous responses have been shown to be associated with
increased levels of fatigue severity and disability (Brooks, Daglish, & Wearden, 2012;
Romano et al., 2009; Schmaling, Smith, & Buchwald, 2000) and recently have been linked
Daily interactions and CFS fluctuations
501
with poorer patient improvement following cognitive behaviour therapy (CBT;
Verspaandonk, Coenders, Bleijenberg, Lobbestael, & Knoop, 2015). The review proposed
that these solicitous responses may promote decreased patient activity (Band et al.,
2015). In turn, reduction in self-reported activity limitation has been shown to mediate the
positive effect of treatment on fatigue (Wearden & Emsley, 2013). However, current
understanding of the relationship between significant other responses and patient CFS/
ME outcomes is based largely upon cross-sectional, patient self-reports of significant
others’ responses, and therefore, alternative methodological techniques are required to
assess the role of significant other responses in symptom maintenance in CFS/ME.
Experience Sampling Methodology (ESM; Csikszentmihalyi & Larson, 1987) utilizes
repeated assessments made within the flow of daily life to assess temporal
associations between variables (Myin-Germeys, Delespaul, & van Os, 2003). As
symptoms associated with chronic conditions such as CFS/ME may fluctuate
considerably over short periods of time (Prins, van der Meer, & Bleijenberg, 2006),
ESM is an appropriate technique to capture relationships between symptom
fluctuations and other factors, and also offers the advantage of addressing some of
the methodological issues associated with symptom reporting in cross-sectional
research (Redelmeier & Kahneman, 1996; Stone & Broderick, 2007; Stone et al.,
2003). ESM is also suitable for examining dyadic interactions (Roche, Pincus, Rebar,
Conroy, & Ram, 2014), offering potential insight into significant other responses
which may vary across contexts within the natural environment (Janicki, Kamarck,
Shiffman, & Gwaltney, 2006; Newton-John & Williams, 2006). Previous studies
examining patient-perceived significant other responses to chronic pain in association
with patient pain intensity (Burns et al., 2013; Sorbi et al., 2006a), and disability
(Sorbi et al., 2006b) have demonstrated that ESM is a feasible methodology for
examining dyadic interactions within chronic conditions such as CFS/ME.
The current study used ESM to investigate significant other responses in association
with patient-reported outcomes within the course of daily functioning, by administering
matched ESM protocols to patients and their closest significant other, each completing the
measures individually and confidentially. It was hypothesized that negative responses
would be associated with increased patient distress and symptom severity. In addition, it
was predicted that significant other solicitous responses would be associated with
increased patient activity limitation and disability. Patient-perceived and significant otherreported responses were investigated. Consistent with the cognitive behavioural model, it
was predicted that the relationship between patient-perceived negative significant other
responses and symptom severity would be mediated by level of patient distress, whilst the
relationship between patient-perceived solicitous significant other responses and levels
of disability would be mediated by patient activity limitation. In addition to exploring
associations between responses and patient outcomes at the same momentary
assessment, lagged analyses were also conducted, to assess the significant other response
reported at the previous momentary assessment in association with change in outcome at
the current assessment (Figure 1).
Method
Design
The ESM protocol was completed by patients and significant others. In addition, patients
completed standardized self-report outcome measures at the start of the study.
502
Rebecca Band et al.
Measured at previous assessment (n-1)
Measured at current assessment (n)
Significant other responses
(Critical and solicitous responses in
the moment)
Patient outcomes
(Symptom severity, distress and
activity limitation in the moment;
patient disability in time elapsed
between beeps)
Significant other responses
(lagged; [n-1])
Time
Figure 1. Depiction of the main effects analyses.
ESM sampling schedule, hardware, and software
Participants followed a typical ESM protocol (Myin-Germeys, van Os, Schwartz, Stone,
& Delespaul, 2001); 10 assessments occurred between typical waking hours of
07.30 am and 10.30 pm, for a period of 6 days. A semi-random sampling schedule
was used; each day of ESM sampling was divided in to ten 90-min periods, and one
assessment was made within each throughout the day. Consecutive beeps occurred
after a minimum of 15 min and a maximum of 3 hr. To ensure all participants
completed assessments within the same time period, access to questions were only
available for 15 min. All alerts and ESM data collection were completed on San
Francisco Android Smartphones using specialist ClinTouch software (Ainsworth et al.,
2013). Data entry was completed using a sliding scale on the touch-sensitive screen
(Ainsworth et al., 2013).
Participants
Participants were recruited from UK CFS/ME services and had received a
specialist clinical diagnosis of CFS/ME, confirmed against the Oxford Criteria
(Sharpe et al., 1991). To be eligible, all patients were required to have a willing
significant other with whom they lived or had at least 10 hr face-to-face contact per
week. All participants had to be aged 16 or above, able to provide fully informed
consent, and have sufficient English comprehension to complete study measures.
Patients were approached to participate at induction to specialist treatment
programmes. A total of 22 dyads consented to participate (approximately 10% of
patients approached) and one dyad self-referred into the study, giving a final sample
of 23 dyads.
Daily interactions and CFS fluctuations
503
Measures
Non-ESM Measures
Patients provided demographic information on illness duration, time of diagnosis, and current
CFS/ME treatment. In addition, validated and frequently used outcome measures assessing
patient fatigue severity and disability were completed prior to the ESM phase of the study.
Patient fatigue and disability. Fatigue was measured using the Chalder Fatigue
Questionnaire (Chalder, Berelowitz, Pawlikowska, Watts, & et al., 1993). It consists of 11
items rated on a 4-point scale (better than usual – much worse than usual; total score
range 0–33); higher scores indicate greater fatigue severity. The scale has been widely
used and well validated in CFS/ME populations (Cella & Chalder, 2010).
The Work and Social Adjustment Scale (WSAS; Marks, 1986) measures functioning
across five areas: Work, home management, social and private leisure activities, and family
relationships. Overall scores are summed from each item (total score range 0–40); higher
scores indicate greater disability. The WSAS has been shown to be a reliable, valid tool for
assessing disability in CFS/ME (Cella, Sharpe, & Chalder, 2011).
Item development. In line with current recommendations, validated non-ESM measures
(Chalder et al., 1993; Marks, 1986; Spence, Moss-Morris, & Chalder, 2005) were used for
item development (Palmier-Claus et al., 2011). The items were phrased in language that
was familiar to patients; positively and negatively worded items were developed to avoid
extreme response bias (Kimhy, Myin-Germeys, Palmier-Claus, & Swendsen, 2012). All
items were rated on a 7-point Likert scale anchored with Not at all to A lot (scored from 1
to 7) and were qualitatively piloted prior to the study commencement to confirm
comprehensibility and acceptability. Momentary items began with the phrase ‘Before the
beep went off I was. . .’ or ‘Right now I am. . .’. STATA (version 11) was used to conduct
preliminary analyses assessing the factor structure of the ESM items for individual
subscales (Myin-Germeys et al., 2001), using the FACTOR command and the ML method
of extraction. The factor solution was determined by identifying the number of factors
with an eigenvalue of greater than one, and individual items with a loading of >0.4 were
included within the subscale. Internal consistencies of subscales were assessed using the
ALPHA command.
Symptom severity: Seven items were developed to assess symptom severity ‘in the
moment’, that is, at the time immediately preceding the alert, and were completed by
patients only. These items reflected core CFS/ME symptoms, including concentration
difficulties, termed ‘mental fog’ by UK sufferers. Items were feeling weak, active, tired,
well, experiencing pain, experiencing mental fog, and being sleepy. Preliminary principal
components analyses indicated that all items loaded on to a one factor solution (a = .79).
Distress: Distress was assessed using a single item (‘feeling distressed’), which was
included with standard items examining participants’ affect at a momentary level.
Activity limitation: Two items were included at the momentary level to assess patient
activity limitation. These items were ‘resting to control my symptoms’ and ‘avoiding
activities that might make my symptoms worse’, completed by patients only. The alpha
coefficient for these items was calculated at a = .80. Whilst it is recognized that
Cronbach’s alpha is not ideal for testing the reliability of a two item scale, it is likely to
produce an underestimate of the true reliability (Eisinga, Grotenhuis, & Pelzer, 2013).
504
Rebecca Band et al.
Disability: Six items were developed to assess patient disability during the time
elapsed since the previous momentary assessment. The phrase ‘Since the last beep I
was able to’ preceded items examining household tasks, socializing, leisure activities,
leaving the house, work, and general activity; a higher score indicated less disability.
All items loaded on to one factor solution (a = .82). These items were completed by
patients only.
Significant other contact and responses: Participants were asked to report on
significant other responses if they indicated dyadic contact at the momentary level.
These were completed confidentially by the patient and the significant other on
individual smartphones. These items were developed to reflect negative and solicitous
responses as defined within the wider literature (Kerns, Turk, & Rudy, 1985; Vaughn
& Leff, 1976). Patients and significant others were asked to rate the extent to which
the significant other was engaging in the behaviour in the moment before the alert
(e.g., ‘Before the beep went off I was doing things for him/her’); responses were
rated from ‘Not at all’ to ‘A lot’. Analyses revealed that significant other responses
loaded on to a two factor structure; the first factor was labelled as negative responses
and included nagging me, irritated with me, and pushing me to do things (a = .92).
The second factor was labelled as solicitous responses and included the following
behaviours: Doing things for me, looking after me, helping me, and checking up on
me (a = .95).
Associations with non-ESM measures
Correlations were conducted between mean levels of ESM reported outcomes (across all
valid beeps) with validated non-ESM outcome measures (Palmier-Claus et al., 2011).
Symptom severity reported during the ESM phase was found to be significantly correlated
with patient-reported fatigue severity (Chalder Fatigue scale; rs = .567, p = .005). High
levels of ESM reported disability were found to be correlated with total levels of WSAS
disability (rs = .538, p = .010).
Procedure
ESM briefing
All participants were given a thorough briefing about the study, which included an ESM
practice session to ensure that participants understood the meaning of ESM items and ESM
rating scales. Researcher and participant contact details were confirmed to maintain
contact during the ESM period. Written informed consent was obtained from all
participants. Patient non-ESM measures were completed prior to the ESM phase of the
study.
Six-day ESM phase
The ESM phase began on the day following the briefing session and the sampling schedule
was synchronized for both patients and significant others. Patients were contacted on Day
2 of the ESM phase to discuss any potential problems and to ensure that participants were
happy to continue with the study. Participants were contacted again 2 days later if
requested. The ESM phase ended after a period of 6 days.
Daily interactions and CFS fluctuations
505
ESM debriefing
After the completion of the ESM phase, all participants were debriefed. All
participants were asked to provide evaluative feedback about their experiences of
the study.
Statistical analysis
Experience Sampling Methodology has a hierarchal structure, whereby measures are
clustered in three levels: Beeps are nested in days which are nested within
participants; therefore, multilevel models were used to test the hypotheses since
these take into account the hierarchical structure. The XTMIXED command in Stata
(StataCorp, 2009) was used for all continuous outcome variables, with a random
intercept for each participant and for each day within participant; betas, 95% CI, and
p-values are reported for all associations between independent and dependent
variables. On the basis of previous research, preliminary analyses were conducted to
examine patient and significant other gender as potential confounders by assessing
their relationship with the outcome variables (Ax, 1999; Tamres, Janicki, & Helgeson,
2002). To test the first set of hypotheses, significant other responses were entered as
predictor variables into the models with patient outcome variables as the dependent
variable for the same momentary assessment (Figure 1). Subsequently, in separate
models, lagged significant other responses reported at the previous assessment
(n
1) were included as predictors with patient outcomes at the current assessment
(n) as the dependent variable. Mediation hypotheses were assessed using a difference
in coefficients approach. Using the XTMIXED command, we fitted a model without
the putative mediator and subsequently including the mediator as a predictor. A
change in the coefficient of the main predictor between these two models can be
interpreted as evidence of mediation and the size of the indirect effect can be
calculated by taking the difference of these coefficients. All analyses were conducted
for both significant other-reported and patient-perceived responses. Finally, intraclass
correlation coefficients (ICCs) were calculated for each of the significant other
response variables, to explore the amount of unexplained variation at each level of
the model. This allows the proportion of total variability in the outcome to be
examined at the person level (i.e., between people), at the day level (i.e., across
different days), and at the beep level (i.e., within the same person and day level, but
across different beeps).
Results
Description of sample
The patient sample (n = 23) ranged in age from 17 to 58, with a mean age of 35.5
(SD = 13.96) years. Twenty (87%) of the sample were female and 21 (91%) were White
British. At the time of recruitment, the median patient illness duration was 5 years
(IQR = 10) and 22 (96%) participants were undergoing specialist CFS/ME treatment
programmes. Three patients (13%) lived alone but nominated the individual with whom
they had the most extensive daily contact as their significant other. Significant others’
(n = 23) ages ranged from 19 to 72, with a mean age of 45 (SD = 13.35) years old.
Thirteen (57%) significant others were female; 11 (48%) were partners, 9 (39%) were
parents, and 3 (13%) were daughters of the patient.
506
Rebecca Band et al.
Patient adherence and retention
None of the 46 participants dropped out before the completion of the study. Three patients
(7%) and two significant others (4%) did not complete the level of valid assessments
(n = 20) traditionally recommended to be retained for analyses (Palmier-Claus et al.,
2011). To exploit all of the available data, the analyses presented here were conducted
including all participants. Patients completed a mean of 38.74 beeps (SD = 14.88), whilst
significant others completed a mean of 34.52 beeps (SD = 14.93). Patients reported a
mean of 18.97 momentary significant other contacts (SD = 11.42), and significant others
reported a mean of 13.26 (SD = 10.90) momentary patient contacts.
Preliminary analysis
Demographic information relating to the variables included within the multilevel model
analyses presented below can be found in Table 1.
Confounding variables
Preliminary analyses indicated that neither patient nor significant other gender significantly predicted patient-reported outcomes on a momentary basis (data not shown). No
further analyses including these variables were therefore conducted.
Effects of negative significant other responses on current patient symptom severity
Both significant other-reported and patient-perceived negative responses were associated
with increased patient symptom severity reported at the same momentary assessment
(Table 2). The negative response variables were then lagged, to reflect the responses
reported at the previous beep (n
1), and regression analyses repeated; these indicated
that there were no significant associations between significant other negative responses at
the previous assessment and current symptom severity.
The mediating effect of patient distress on the relationship between negative significant
other responses and current patient symptom severity
Analyses revealed that negative responses as reported by the patient and significant others
were also associated with increased patient distress on a momentary basis. To assess the
Table 1. Demographic information for predictor and outcome variables included within the multilevel
model analyses
No of observations Min, Max Mean (SD) subscale score
Symptom severity
Distress
Disability
Activity limitation
Patient-reported SO solicitous responses
Patient-reported SO negative responses
SO reported solicitous responses
SO reported negative responses
Note. SO = significant other.
894
894
894
894
894
894
809
809
0, 41
0, 7
0, 7
0, 7
0, 7
0, 7
0, 7
0, 7
16.74 (13.19)
2.07 (1.64)
2.79 (1.46)
3.94 (2.09)
1.36 (1.84)
0.78 (1.11)
1.00 (1.60)
0.54 (0.87)
Daily interactions and CFS fluctuations
507
Table 2. The association between significant other responses and patient outcomes in momentary and
lagged (n
1) analyses
Patient distress
Predictor variables
b
Significant other negative responses
Momentary
.173
Lagged (n
1)
.043
Patient-perceived negative responses
Momentary
.188
Lagged (n
1)
.026
Patient symptom severity
SE
p
.059
.060
.003
.48
.041
.048
<.001
.59
b
SE
p
2.295
.489
.427
.467
.001
.30
.505
621
.149
.357
.001
.082
Patient activity limitation
b
Significant other solicitous responses
Momentary
.046
Lagged (n
1)
.096
Patient-perceived solicitous responses
Momentary
.080
Lagged (n
1)
.069
Patient disability
SE
p
b
SE
p
.053
.058
.38
.098
.034
.007
.031
.034
.27
.85
.023
.044
<.001
.11
.094
.003
.038
.025
.013
.92
p < .05 is in boldface.
potential mediating effect of distress, the multilevel model predicting momentary patient
symptom severity was recalculated including the potential mediating variable, distress, in
the model. When patient distress, significant other-reported and patient-perceived
negative responses were examined in a single model, only patient momentary distress
remained as a significant predictor, suggesting that the effect of significant other negative
responses on patient symptom severity is explained by increased levels of momentary
patient distress (Table 3).
Effects of solicitous significant other responses on current patient disability
Contrary to study hypotheses, patient-perceived solicitous responses were associated
with decreased levels of patient-reported disability at the concurrent momentary
assessment. However, significant other-reported solicitous responses were not found to
be associated with momentary reports of patient disability (Table 2).
The relationship between activity limitation, solicitous significant other responses, and
current patient disability
Increased levels of self-reported patient activity limitation were significantly associated
with higher levels of patient-perceived solicitous responses at the same momentary
assessment but not significant other-reported solicitous responses. In addition, patient
activity limitation was found to significantly predict increased patient disability at the
momentary assessment (Table 2). When both variables were included in subsequent
multilevel models, patient-perceived solicitous responses and activity limitation were
identified as independent predictors of patient disability on a momentary level (Table 3),
508
Rebecca Band et al.
Table 3. Single models combining all predictor variables on patient outcomes at the momentary level
Patient symptom severity
Predictor variables
Model 1
Patient distress
Model 2
Significant other negative responses
Patient-perceived negative responses
Patient distress
b
SE
p
.708
.118
<.001
.390
.165
.633
.222
.185
.139
.080
.37
<.001
Patient disability
b
Model 1
Activity limitation
Model 2
Patient-perceived solicitous responses
Activity limitation
SE
p
.196
.019
<.001
.099
.204
.022
.019
.001
<.001
p < .05 is in boldface.
indicating that on a momentary basis, patient activity limitation did not appear to mediate
the association between solicitous responses and patient disability.
The variability of patient-reported outcomes across the different levels of data
The ICC analyses indicate that the majority of the unexplained variation in patient
symptom severity and activity limitation occurred at the beep level (i.e., between
assessments, within the same patient, and across the same day). However, patient levels of
distress and disability indicate much higher levels of variation between individual
participants (Table S1). For example, when examining significant other negative
responses as the predictor, 66% of the unexplained variation in symptom severity is at
the beep level. However, whilst distress also varied most (49%) at the beep level, 35% of
the variance was identified at the person level, suggesting that distress varies more
between people than symptom severity.
Discussion
The current study aimed to examine two types of significant other responses
hypothesized to be important in association with patient-reported CFS/ME outcomes
within the context of daily life. In line with study hypotheses, significant other negative
responses were associated with increases in patient-reported symptom severity and
distress at the same momentary assessment. The relationship between negative significant
other responses and increased patient depression has been documented cross-sectionally
(Romano et al., 2009; White et al., 2006) and patient depression has also been identified
as predictive of poorer patient illness and treatment outcomes (Bentall et al., 2002;
Tamres et al., 2002). A recent review of the existing evidence suggested that significant
other negative responses may be important for patient illness outcomes by increasing
Daily interactions and CFS fluctuations
509
levels of patient depression (Band et al., 2015) supported by longitudinal evidence
demonstrating the mediational role of increased patient depression between significant
other criticism and fatigue severity (Band et al., 2014). The findings presented here
further support the proposed interpersonal process (Band et al., 2015) and extend the
current literature by demonstrating the associations between these variables on a
momentary basis.
It was also hypothesized that significant other solicitous responses would be
associated with increased patient disability and activity limitation. Patient-perceived
solicitous responses were observed to be associated with increased activity limitation
at the same momentary assessment, in line with study hypotheses. Previous research
has indicated that activity limitation may be an important factor mediating the effect
of treatment programmes in reducing fatigue (Wearden & Emsley, 2013), and the
current analyses also suggest that patient-perceived solicitous responses have an
independent effect on momentary levels of patient activity limitation. However, the
association between patient-perceived solicitous significant other responses and
patient disability was in the opposite direction to that predicted, with increased
solicitousness associated with reduced patient-reported disability at the same
momentary assessment. This finding would at first sight to appear be inconsistent
with the previous literature where, cross-sectionally, and using different methodologies, increased patient disability and fatigue have been associated with increased
significant other solicitous responses (Brooks et al., 2012; Romano et al., 2009;
Schmaling et al., 2000). The interpretation of our finding is limited by the
correlational associations of these variables. Possibly the finding indicates that when
patients perceive themselves as more able to engage in activities such as work and
social activities, they also report their significant other as engaging in more solicitous
behaviours (e.g., helping or looking after them). Equally, it is also plausible that
significant others respond in a more solicitous way when patients are limiting their
activity, leading to a perceived reduction in patient disability.
The items developed to assess patient disability showed a significant, moderate
correlation with the WSAS measure of disability, suggesting that patients reporting greater
levels of momentary disability also reported higher levels of ongoing disability. However,
the inconsistency between our findings with respect to solicitous significant other
responding and patient disability and those previously reported may reflect the
methodological differences between measuring variables within an ESM paradigm and
using traditional cross-sectional questionnaire reports; the ESM data may potentially
reflect more state-like reports of disability, whereas cross-sectional reports may reflect
more stable, global perceptions of disability. ESM data offer the potential to examine
temporal relationships between variables as they occur, or between current events (e.g.,
significant other responses) and subsequent patient outcomes. On a momentary level,
patient-perceived solicitous responses may be experienced as helpful and facilitative by
patients, but repeated solicitous exchanges with significant others may impact upon
patient outcomes differently over time, as the wider cross-sectional literature would
suggest. To address this possibility, future research examining the impact of significant
other behaviour might usefully examine those dyads where the significant other has a
persistent and frequent response style (i.e., negative or solicitous) compared with dyads
where behavioural interactions with the patient are more variable.
Whilst consistent associations were noted for patient-perceived and significant otherreported negative responses, it is worth noting that significant other-reported solicitous
responses were not found to be significantly associated with patient disability or activity
510
Rebecca Band et al.
limitation, as expected. The inclusion of significant other-reported responses was
beneficial for examining these processes from both a patient and significant other
viewpoint, and suggests that the respondent (i.e., patient or significant other) is an
important factor that has received little attention in the previous literature.
However, a second, quite different interpretation of our unexpected finding is that
solicitous behaviour on the part of significant others actually was leading to reductions in
patient disability and that the inconsistency with previous literature is because the
solicitousness which we measured was somewhat different from that which has been
measured in other studies. Solicitous responding is likely to be a compound construct, and
it is possible that some aspects of what might be labelled solicitousness, for example,
providing encouragement for appropriate levels of activity, may be helpful for the
patient’s recovery, whilst other aspects of solicitousness, such as overprotection, may be
unhelpful.
Many of the items utilized within the current study were developed specifically, and as
a result, it is possible that solicitous response items included within the study do not
accurately reflect either the same type or same level of solicitous behaviours typically
found to associate with poorer patient outcomes. We further recognize that other
constructs utilized within our study, such as distress, are related but not identical to those
previously reported in the literature (such as depression; Mead, 2002). Furthermore, the
items used to assess activity limitation within the current study required patients to report
on their activity levels and make a judgement on why they were limiting their activity, if
they were doing so at the momentary assessment. These items limit the interpretation of
the associations between solicitous responses, activity limitation, and disability; particularly as we are unable to account for instances where patients may have been resting but
were not consciously doing so to control symptoms, for example. Whilst inherently
subjective sensations such as fatigue or pain cannot be objectively measured, the
development and inclusion in future studies of objective measures such as actigraphy may
help overcome some of these difficulties.
There are a number of limitations associated with the current study that need to
be acknowledged. As noted above, the temporal associations identified between
variables of interest are limited by their correlational nature. Therefore, it is not
possible to infer causal relationships from ESM data, and as a result, alternative
explanations for these observations must be explored. For example, it is possible that
increased patient symptom severity at the momentary level elicited negative
significant other responses. The lack of associations between significant other
responses and patient outcomes at the subsequent momentary assessment is
surprising given the previous literature and suggests that the impact of significant
other responses on patient outcomes is, at a momentary level at least, fairly transitory.
It is also possible that patient enrolment in specialist CFS/ME treatment programmes
prior to entering the study may have impacted upon the relationship between patient
illness experiences and significant other responses. Patients may have been at
different stages of their treatment programmes; we do not have data regarding the
stage of treatment that each patient had reached. Future studies may usefully combine
ESM techniques with ongoing treatment programmes to analyse processes of change.
Additionally, a final limitation is the reliance upon participant self-report in generating
momentary data, particularly in relation to potentially sensitive questions such as
significant other responses. However, confidential electronic data collection ensured
that all participants’ data entry remained private and was inaccessible to either
participant following assessment completion.
Daily interactions and CFS fluctuations
511
Conclusions
The novel findings from this study demonstrate that significant other responses are
important for patients’ day-to-day experience of CFS/ME. In particular, the results indicate
that targeting significant other negative responses may be beneficial for reducing
temporary increases in patient experience of symptom severity and distress. The
association between solicitous responses (as perceived by the patient) and patient activity
limitation is important, given the reduction in activity limitation is associated with
decreased fatigue severity during patient treatment programmes (Wearden & Emsley,
2013). The results also indicate that perceived helpful responses may be important in
facilitating patient perception of increased ability to participate in daily activities. Given
that that cognitive behavioural interventions aim to address dysfunctional beliefs about
the relationships between symptom experience and activity, it is important to know
whether these beliefs and the behaviour patterns they engender are being reinforced by
significant others (Wiborg, Knoop, Stulemeijer, Prins, & Bleijenberg, 2010). Future
research should seek to examine which specific significant other responses elicit
increased levels of patient activity, and which associate most strongly with unfavourable
activity management strategies, such as activity limitation. We suggest that it is
particularly important to explore further the nature and role of solicitous significant
other responding, as this may inform potentially helpful significant other focused
interventions.
Acknowledgements
This study was supported by a PhD studentship awarded to the first author by the UK Economic
and Social Research Council (ESRC) whilst at the University of Manchester. The research team
acknowledges the support of the National Institute for Health Research, through the
Comprehensive Clinical Research Network.
Conflict of interest
All authors declare no conflict of interest.
References
Ainsworth, J., Palmier-Claus, J. E., Machin, M., Barrowclough, C., Dunn, G., Rogers, A., . . . Lewis, S.
(2013). A comparison of two delivery modalities of a mobile phone-based assessment for serious
mental illness: Native smartphone application vs text-messaging only implementations. Journal
of Medical Internet Research, 15(4), e60. doi:10.2196/jmir.2328
Ax, S. (1999). Coping differences between chronic fatigue syndrome sufferers and their carers.
Journal of Chronic Fatigue Syndrome, 5(2), 27–62. doi:10.1300/J092v05n02_03
Band, R., Barrowclough, C., & Wearden, A. (2014). The impact of significant other expressed
emotion on patient outcomes in chronic fatigue syndrome. Health Psychology, 33, 1092–1101.
doi:10.1037/hea0000086
Band, R., Wearden, A., & Barrowclough, C. (2015). Patient outcomes in association with significant
other responses to chronic fatigue syndrome: A systematic review of the literature. Clinical
Psychology: Science and Practice, 22(1), 29–46. doi:10.1111/cpsp.12093
Bentall, R. P., Powell, P., Nye, F. J., & Edwards, R. H. (2002). Predictors of response to treatment for
chronic fatigue syndrome. British Journal of Psychiatry, 181, 248–252. doi:10.0.4.168/
bjp.181.3.248
512
Rebecca Band et al.
Brooks, J., Daglish, J., & Wearden, A. (2013). Attributions, distress and behavioural responses in the
significant others of people with chronic fatigue syndrome. Journal of Health Psychology, 18,
1288–1295. doi:10.1177/1359105312464670
Burns, J. W., Peterson, K. M., Smith, D. A., Keefe, F. J., Porter, L. S., Schuster, E., & Kinner, E. (2013).
Temporal associations between spouse criticism/hostility and pain among patients with chronic
pain: A within-couple daily diary study. Pain, 154, 2715–2721.
Castell, B. D., Kazantzis, N., & Moss-Morris, R. E. (2011). Cognitive behavioral therapy and graded
exercise for chronic fatigue syndrome: A meta-analysis. Clinical Psychology: Science and
Practice, 18, 311–324. doi:10.1111/j.1468-2850.2011.01262.x
Cella, M., & Chalder, T. (2010). Measuring fatigue in clinical and community settings. Journal of
Psychosomatic Research, 69(1), 17–22. doi:10.1016/j.jpsychores.2009.10.007
Cella, M., Sharpe, M., & Chalder, T. (2011). Measuring disability in patients with chronic fatigue
syndrome: Reliability and validity of the Work and Social Adjustment Scale. Journal of
Psychosomatic Research, 71(3), 124–128. doi:10.1016/j.jpsychores.2011.02.009
Chalder, T., Berelowitz, G., Pawlikowska, T., Watts, L., Wessely, S., Wright, D., Wallace, E.P. (1993).
Development of a fatigue scale. Journal of Psychosomatic Research, 37(2), 147–153.
doi:10.1016/0022-3999%2893%2990081-P
Cordingley, L., Wearden, A., Appleby, L., & Fisher, L. (2001). The Family Response Questionnaire: A
new scale to assess the responses of family members to people with chronic fatigue syndrome.
Journal of Psychosomatic Research, 51, 417–424. doi:10.1016/S0022-3999%2801%2900193-3
Csikszentmihalyi, M., & Larson, R. (1987). Validity and reliability of the experience-sampling
method. Journal of Nervous and Mental Disease, 175, 526–536. doi:10.1097/00005053198709000-00004
Deary, V., Chalder, T., & Sharpe, M. (2007). The cognitive behavioural model of medically
unexplained symptoms: A theoretical and empirical review. Clinical Psychology Review, 27,
781–797. doi:10.1016/j.cpr.2007.07.002
Dickson, A., Knussen, C., & Flowers, P. (2007). Stigma and the delegitimation experience: An
interpretative phenomenological analysis of people living with chronic fatigue syndrome.
Psychology & Health, 22, 851–867. doi:10.1080/14768320600976224
Eisinga, R., Grotenhuis, M., & Pelzer, B. (2013). The reliability of a two-item scale: Pearson,
Cronbach, or Spearman-Brown? International Journal of Public Health, 58, 637–642.
doi:10.1007/s00038-012-0416-3
Fukuda, K., Straus, S. E., Hickie, I., Sharpe, M. C., Dobbins, J. G., Komaroff, A. L., & I. C. F. S. S.
Group (1994). The Chronic Fatigue Syndrome: A comprehensive approach to its definition
and study. Annals of Internal Medicine, 121, 953–959. doi:10.7326/0003-4819-121-12199412150-00009
Janicki, D. L., Kamarck, T. W., Shiffman, S., & Gwaltney, C. J. (2006). Application of ecological
momentary assessment to the study of marital adjustment and social interactions during daily
life. Journal of Family Psychology, 20(1), 168–172.
Kerns, R. D., & Rosenberg, R. (1995). Pain-relevant responses from significant others: Development
of a significant-other version of the WHYMPI scales. Pain, 61, 245–249. doi:10.1016/0304-3959
(94)00173-c
Kerns, R. D., Turk, D. C., & Rudy, T. E. (1985). The West Haven-Yale Multidimensional Pain
Inventory (WHYMPI). Pain, 23, 345–356. doi:10.1016/0304-3959%2885%2990004-1
Kimhy, D., Myin-Germeys, I., Palmier-Claus, J., & Swendsen, J. (2012). Mobile assessment guide for
research in schizophrenia and severe mental disorders. Schizophrenia Bulletin, 38, 386–395.
Marks, I. (1986). Behavioural psychotherapy. Bristol, UK: John Wright.
Mead, D. E. (2002). Marital distress, co-occuring depression and marital therapy: A review. Journal
of Marital and Family Therapy, 28, 299–312.
Myin-Germeys, I., Delespaul, P., & van Os, J. (2003). The Experience Sampling Method in psychosis
research. Current Opinion in Psychiatry, 16(Suppl 2), S33–S38. doi:10.1097/00001504200304002-00006
Daily interactions and CFS fluctuations
513
Myin-Germeys, I., van Os, J., Schwartz, J. E., Stone, A. A., & Delespaul, P. A. (2001). Emotional
reactivity to daily life stress in psychosis. Archives of General Psychiatry, 58, 1137–1144.
doi:10.1001/archpsyc.58.12.1137
Newton-John, T. R., & Williams, A. (2006). Chronic pain couples: Perceived marital interactions and
pain behaviours. Pain, 123(1–2), 53–63. doi:10.1016/j.pain.2006.02.009
Palmier-Claus, J., Myin-Germeys, I., Barkus, E., Bentley, L., Udachina, A., Delespaul, P., . . . Dunn, G.
(2011). Experience sampling research in individuals with mental illness: Reflections and
guidance. Acta Psychiatrica Scandinavica, 123(1), 12–20. doi:10.1111/j.16000447.2010.01596.x
Prins, J. B., van der Meer, J. W. M., & Bleijenberg, G. (2006). Chronic fatigue syndrome. The Lancet,
367, 346–355. doi:10.1016/S0140-6736(06)68073-2
Redelmeier, D. A., & Kahneman, D. (1996). Patients’ memories of painful medical treatments: Realtime and retrospective evaluations of two minimally invasive procedures. Pain, 66(1), 3–8.
doi:10.1016/0304-3959(96)02994-6
Roche, M. J., Pincus, A. L., Rebar, A. L., Conroy, D. E., & Ram, N. (2014). Enriching psychological
assessment using a person-specific analysis of interpersonal processes in daily life. Assessment,
21(5), 515–528. doi:10.1177/1073191114540320
Romano, J. M., Jensen, M. P., Schmaling, K. B., Hops, H., & Buchwald, D. S. (2009). Illness behaviors
in patients with unexplained chronic fatigue are associated with significant other responses.
Journal of Behavioral Medicine, 32, 558–569. doi:10.1007/s10865-009-9234-3
Schmaling, K. B., Smith, W. R., & Buchwald, D. S. (2000). Significant other responses are associated
with fatigue and functional status among patients with chronic fatigue syndrome.
Psychosomatic Medicine, 62, 444–450.
Sharpe, M. C., Archard, L. C., Banatvala, J. E., Borysiewicz, L. K., Clare, A. W., David, A., . . . Lane, R. J.
(1991). A report-chronic fatigue syndrome: Guidelines for research. Journal of the Royal Society
of Medicine, 84(2), 118–121.
Sorbi, M. J., Peters, M. L., Kruise, D. A., Maas, C. J., Kerssens, J. J., Verhaak, P. F., & Bensing, J. M.
(2006a). Electronic momentary assessment in chronic pain I: Psychological pain responses as
predictors of pain intensity. The Clinical Journal of Pain, 22(1), 55–66. doi:10.1097/
01.ajp.0000148624.46756.fa
Sorbi, M. J., Peters, M. L., Kruise, D. A., Maas, C. J., Kerssens, J. J., Verhaak, P. F., & Bensing, J. M.
(2006b). Electronic momentary assessment in chronic pain II: Pain and psychological pain
responses as predictors of pain disability. The Clinical Journal of Pain, 22(1), 67–81.
doi:10.1097/01.ajp.0000148625.84874.48
Spence, M., Moss-Morris, R., & Chalder, T. (2005). The Behavioural Responses to Illness
Questionnaire (BRIQ): A new predictive measure of medically unexplained symptoms
following acute infection. Psychological Medicine, 35, 583–593. doi:10.1017/
S0033291704003484
StataCorp (2009). Stata statistical software: Release 11. College Station, TX: Author.
Stone, A. A., & Broderick, J. E. (2007). Real-time data collection for pain: Appraisal and current status.
Pain Medicine, 8(Suppl 3), S85–S93. doi:10.1111/j.1526-4637.2007.00372.x
Stone, A. A., Broderick, J. E., Schwartz, J. E., Shiffman, S., Litcher-Kelly, L., & Calvanese, P. (2003).
Intensive momentary reporting of pain with an electronic diary: Reactivity, compliance, and
patient satisfaction. Pain, 104(1–2), 343–351. doi:10.1016/s0304-3959(03)00040-x
Surawy, C., Hackmann, A., Hawton, K., & Sharpe, M. (1995). Chronic fatigue syndrome: A cognitive
approach. Behaviour Research and Therapy, 33, 535–544. doi:10.1016/0005-7967%2894%
2900077-W
Tamres, L. K., Janicki, D., & Helgeson, V. S. (2002). Sex differences in coping behavior: A metaanalytic review and an examination of relative coping. Personality & Social Psychology Review
(Lawrence Erlbaum), 6(1), 2–30.
Vaughn, C., & Leff, J. (1976). The measurement of expressed emotion in the families of psychiatric
patients. British Journal of Social & Clinical Psychology, 15(2), 157–165.
514
Rebecca Band et al.
Verspaandonk, J., Coenders, M., Bleijenberg, G., Lobbestael, J., & Knoop, H. (2015). The role of the
partner and relationship satisfaction on treatment outcome in patients with chronic fatigue
syndrome. Psychological Medicine FirstView, 45(11), 2345–2352. doi:10.1017/S003329171
5000288
Wearden, A. J., Dunn, G., Dowrick, C., & Morriss, R. (2012). Depressive symptoms and pragmatic
rehabilitation for chronic fatigue syndrome. The British Journal of Psychiatry, 201, 227–232.
doi:10.1192/bjp.bp.111.107474
Wearden, A. J., & Emsley, R. (2013). Mediators of the effects on fatigue of pragmatic rehabilitation for
chronic fatigue syndrome. Journal of Consulting and Clinical Psychology, 81, 831–838.
doi:10.1037/a0033561
White, P., Goldsmith, K., Johnson, A., Potts, L., Walwyn, R., Decesare, J., . . . Cox, D. (2011).
Comparison of adaptive pacing therapy, cognitive behaviour therapy, graded exercise therapy,
and specialist medical care for chronic fatigue syndrome (PACE): A randomised trial. Lancet,
377, 823–836.
White, K., Lehman, D. R., Hemphill, K. J., Mandel, D. R., & Lehman, A. M. (2006). Causal attributions,
perceived control, and psychological adjustment: A study of chronic fatigue syndrome. Journal
of Applied Social Psychology, 36(1), 75–99. doi:10.1111/j.0021-9029.2006.00004.x
Wiborg, J., Knoop, H., Stulemeijer, M., Prins, J., & Bleijenberg, G. (2010). How does cognitive
behaviour therapy reduce fatigue in patients with chronic fatigue syndrome? The role of physical
activity. Psychological Medicine: A Journal of Research in Psychiatry and the Allied Sciences,
40, 1281–1287. doi:10.1017/S0033291709992212
Received 10 April 2015; revised version received 11 November 2015
Supporting Information
The following supporting information may be found in the online edition of the article:
Table S1. Intraclass correlation coefficients (ICC) for potential significant other
response predictor variables.